Machine Learning Engineer | Researcher | M.S. CS @ UIC (2024–2026)
Federated Learning • LLMs • Cloud-Native Systems • MLOps

Email • LinkedIn • GitHub

Engineer. Researcher. Builder.
I design and deploy AI systems that scale, from intelligent applications to robust ML infrastructure.
When I’m not working on models or systems, you’ll probably find me on the cricket field or in the boxing ring — still chasing performance, just in a different arena.

Sep 2024 – Present, Chicago, IL

• Designed federated learning frameworks with efficient updates (MSB–LSB, adaptive gradient freezing), cutting overhead by 80%.
• Implemented early-exit neural architectures for adaptive inference on edge devices, improving energy efficiency while maintaining accuracy.
• Researched and deployed Weightless Neural Networks (WNNs) for healthcare and biosignal processing.
• Built active learning modules with uncertainty sampling, accelerating convergence under limited labels.

Jan 2022 – Jul 2024, Mumbai, India

• Designed and deployed anomaly detection models (Random Forest, Isolation Forest, Transformers) with 97% precision in production.
• Built scalable real-time ML pipelines for log and behavior analytics, integrating 30+ data sources and cutting latency by 40%.
• Applied AI-driven threat modeling and root-cause analysis aligned with MITRE ATT&CK for improved fraud/security insights.

Mar 2021 – Jun 2022, Mumbai, India

• Developed and deployed ML models for anomaly detection and predictive analytics, boosting reliability by 30%.
• Built scalable data pipelines and feature engineering workflows, reducing model training time by 25%.
• Integrated models into production with REST APIs and Docker for seamless scalability.

• Programming: Python • C++ • Java • SQL • C#
• ML/AI: Deep Learning • LLMs • Transformers • GANs • NLP • Model Optimization • Adversarial Robustness • Explainable AI
• Frameworks: PyTorch • TensorFlow • Keras • Scikit-learn • Hugging Face • Pandas • NumPy
• Cloud/DevOps: AWS • Azure • Docker • Kubernetes • Jenkins • CI/CD • Cloud-Native Inference • Model Serving
• Databases: MySQL • MongoDB • PostgreSQL

• Converso (LLM Chatbot Platform): Production-grade chatbot (LangChain + OpenAI APIs + FAISS) deployed on AWS with Docker/Kubernetes, scaled to 50K+ daily queries with 99.9% uptime.
• RecoTrack (Real-Time Recommendation Engine): Personalized recommendation engine (XGBoost + collaborative filtering), achieving an 18% CTR lift and integrated with React front-end.
• FraudShield (Transaction Monitoring): Streaming fraud detection pipeline (Kafka + PySpark + Random Forest) with 95% accuracy and sub-200ms response time in AWS.

• University of Illinois at Chicago — M.S. in Computer Science (2024–2026)
• MIT - World Peace University, Pune — B.Tech in Computer Engineering (2018–2022, CGPA: 8.86)

• ATM-Net: Adaptive Termination and Multi-Precision Neural Networks — HPCA 2025
• SenGuard: In-Sensor Privacy-Preserving Processing for Smart Imaging — GLSVLSI 2025
• OrganoSense: Biosignal Neural Processing via Organic Circuits — MWSCAS 2025